Lung Cancer:GEB has been a pioneer in studies aimed at identifying a genetic component to common cancers using epidemiological study designs and we currently have a large lung cancer case-control study (see below- EAGLE, Environment and Genetics in Lung Cancer Etiology) with a major behavior and tissue component. Early work emphasized polymorphic genes (e.g., CYP2D6, GSTM1) that activate or detoxify carcinogens and thereby alter an individual's susceptibility to cancer when exposed to a specific agent however emphasis has now shifted to anonymous searches for candidate genes using high technology approaches. Lung cancer is a logical focus for these studies, because the environmental agent is well characterized (i.e., tobacco smoking), although only a minority of heavy smokers develop the disease. Furthermore, somatic gene mutations are well documented in lung tumors and relatives of people who develop the cancer are also at increased risk. To better understand the genetic component to smoking-related cancer we have conducted case-control studies of lung cancer with extensive biospecimen collection; studied determinants of the cancer in nonsmokers; studied cancer in different geographic, occupational, and ethnic groups; and examined tumor mutations in consecutive surgical cases. We are currently involved in studies to evaluate whether genes contribute to precursor conditions (i.e. MBL in CLL and MGUS in Waldenstroms) or to the key exposure (i.e. smoking, caffeine, alcohol). Families with lung and other smoking-related cancers have been studied, although suitable large, multi generation families are challenging to accrue because of the death toll for smoking-related diseases. Integrated Studies: In our single largest study, EAGLE involving over 2000 newly diagnosed lung cancer cases and 2000 population-based controls, we are investigating the genetic determinants of lung cancer and smoking in a multi-hospital, population-based case-control study of lung cancer with sufficient power to detect gene-environment interactions, biospecimen collection to allow interdisciplinary study of biomarkers from normal and neoplastic tissue, tissue collection from lung cancer surgical cases, and the planned study of genetic determinants of smoking in controls. We conduct studies in the Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Study, a cohort study of 70,000 individuals in which large numbers of lung cancer cases will be identified over the next few years. Previous studies have investigated possible infections agents (Chlamydia pneumoniae and H pylori) in lung cancer, functional DNA repair, and the role of tobacco-related carcinogens (NNAL) in lung cancer. Using our GWAS on lung cancer and the smoking phenotype on over 5500 subjects from EAGLE and PLCO, we recently described the relationship of the Fagerstrom Index (Time to First Cigarette) on risk of lung cancer, incorporating data from both EAGLE and PLCO. This study was published last year. A follow-up on TTFC and COPD was conducted this year and a follow-up study is being conducted in NLST to evaluate the impact of TTFC on risk. Other malignanciesWe intend to incorporate treatment and outcome information in studies of breast, GI and skin malignancies so that these areas can be investigated as technologies to query pharmacogenomic variation become more available. Tobacco and other exposures with a component of dependencyIn the past we investigated the hypothesis of a genetic component to smoking and obesity using candidate gene approaches, e.g. publishing results implicating polymorphisms affecting dopamine regulation and receptor stimulation as candidates for influencing genetic susceptibility to cigarette smoking as well as having a role in energy balance. More recently, we have conducted GWAS of smoking, alcohol, BMI, and caffeine. Other pharmacogenetic hypothesesIn the past, we used the Collaborative Perinatal Project to investigate the influence of genetic differences in phenytoin disposition in mothers on the presence of craniofacial abnormalities in the child. We identified 174 pregnancies in 155 women who used phenytoin throughout their pregnancy and identified 19 children with craniofacial abnormalities. Maternal EPHX genotype showed a significant protective association with craniofacial abnormalities in the child while CYP2C9 genotype was unrelated to fetal endpoints. COPD is another example of a condition associated with malignancy which likely has both shared and independent genetic features with lung cancer. This year we investigated TTFC in relation to COPD and will follow this by looking at the role of pharmacogenetic variation. Future work will incorporate metabolomic and microbiome markers. We are considering the application of mendelian randomization approaches to refine understanding of causality. New data expected from Oncoarray and